To Cope with North Korea's Drone Threat

Oh Dae-kun (right), a senior researcher at the Collaborative Robot Convergence Research Center, pose for a photo shoot with members of his research team after testing the newly developed radar at an unmanned aerial flight test center.

A Korea-U.S. research team has developed artificial intelligence-based radar to cope with the drone threat from North Korea. The radar is powerful enough to detect an ultra-small drone over three kilometers away.

The Daegu Gyeongbuk Institute of Science and Technology (DGIST) announced on July 16 that a team led by Oh Dae-kun, a senior researcher at the Collaborative Robot Convergence Research Center of the DGIST, has developed the drone detection radar system with a team led by professor Kim Young-wook of California State University Fresno. The drone detection radar system is expected to significantly contribute to the enhancement of South Korea’s defense readiness.

The system uses active electronically scanned array (AESA) radar technology to identify a drone 55 centimeters long, 55 centimeters wide and 40 centimeters tall over three kilometers away. This means that the system is comparable to Israel's RADA which is assessed as the world's top-rated radar.

The AESA radar is an anti-air detection system that can simultaneously identify multiple objects in various directions in the sky with multiple eyes like those of bees. Unlike mechanical radar that detects surroundings by rotating a dish-shaped antenna, the AESA radar is loaded with small antenna chips as small as a nail.

The system is also powered by next-generation deep learning technology for AI. The cognitive technology for radar is based on generative adversarial networks (GANs) and a machine learning algorithm that allows AI to easily identify moving targets with only a small amount of data.

The Korean and U.S. researchers expanded the possibility of Korea’s technological independence as they developed the technology with small Korean companies, the DGIST said.

The research results were published in the online edition of IEEE Geoscience and Remote Sensing Letters, an international journal, on June 22.

Copyright © BusinessKorea. Prohibited from unauthorized reproduction and redistribution